GLOS is a multi-phase digital humanities project exploring the conceptual structures and geographic distribution of motifs, themes, and relationships in global vernacular literature, particularly folktales and creation myths.
In this initial phase, the Thompson Motif Index and Uther Tale Type Index were digitized, organized in a database, and embedded using NLP techniques. Two prototype tools were built: an ATU/TMI Cross-reference explorer, and a Concept Matcher that retrieves nearest-neighbor motifs or tale types from any input text.
This ongoing phase focuses on extracting and modeling conceptual content from 124 creation myths sourced from Barbara Sproul’s Primal Myths. Using JSON-LD representations, ontological classes, and conceptual profiling, aided by Large Language Model (LLM) technology, this phase will supports similarity analysis, clustering, and eventual geographic visualization.
➡️ README_myths